Using hamster as a model for wound healing, we have previously shown that, unlike the sequential expression of transforming growth factors alpha and beta1 by eosinophils in skin wounds, eosinophils infiltrating into oral wounds were found to synthesize only TGF-beta1 mRNA (Tyler et al., J Dent Res. 72: 291, Abst. 1500, 1993). We hypothesized that the lack of TGF-alpha expression by eosinophils in oral wounds is due to the presence of epidermal growth factor (EGF) and/or TGF-alpha in hamster saliva. To test this hypothesis, thirty-four hamster were divided into three groups. Group 1 was sham operated (G1), group 2 and 3 were sialoadenectomized (G2&G3) ( to eliminate salivary EGF) 14 days prior to creating ~8mm circular mucosal wound overlying the messeter muscle. The third group of hamsters (G3) was used to determine, among those sialoadenectomized hamsters, whether EGF-supplemented water could restore the normal healing process. Oral wounds from 2 animals in each group were harvested every other day until day 10. Each harvested wound was processed for histology to examine eosinophil infiltration and for in-situ hybridization to examine the expression of TGF-alpha mRNA, using hamster-specific 35S-labeled TGF-alpha riboprobe. The rate of mucosal closure was monitored clinically and analyzed by planimetry. The result showed that clinical wound closure among the three groups varied from 6-8 days. Interestingly, G3 hamsters healed the fastest (day 6), although this difference among the groups was not significant at complete wound closure. This is in contrast to previous reports that lack of salivary EGF retarded mucosal wound healing. At the cellular level, eosinophils infiltrated into oral wounds of all three groups. Contrary to the sham-operated and G3 groups, where the infiltrated eosinophils were not expressing TGF-alpha mRNA, the sialoadenectomized group (G2), i.e. without EGF-supplemented water, eosinophils were found to express TGF-alpha. We hypothesize that the restoration of eosinophil expression of TGF-alpha in oral wounds in G2 hamsters is responsible for the comparable healing rates of the G2 and G3 group. We further propose that salivary TGF-alpha/EGF and eosinophils-derived TGF-alpha operate in a complementary/compensatory fashion in hamster oral wound healing. Key Words: Eosinophil, Wound Healing, TGFs, EGF, Hamster, Sialoadenectomy